The central goal of insulin replacement therapy in the treatment of diabetes mellitus (DM) is tight control of blood glucose concentrations. Clinical trials, including the landmark Diabetes Control &Complications Trial (DCCT) and follow-up Epidemiology of Diabetes Interventions and Complications Study (DCCT/EDIC), have documented in Type 1 DM the benefit of tight glycemic control in reducing the risk and delaying the progression of long-term cardiovascular and microvascular complications, including diabetic neuropathy, retinopathy, and renal disease. Similar trends are observed in Type 2 DM. These data have motivated extensive efforts to develop an implantable insulin pump for regulated peritoneal deliver. To date, however, implantable pumps remain experimental due to the problem of pump occlusion associated with insulin fibrillation. The objective of this application is to design an optimal insulin analog and insulin formulation for safe and effective use in an implantable pump. The proposed design strategy, based on general scientific principles of protein structure, employs single-chain insulin analogs. Despite their many theoretical advantages and promising clinical trials, implantable insulin pumps are presently not approved by the FDA. A major problem is posed by the limited stability of present insulin formulations as stored for 1-3 months within the pump reservoir at 37 oC within the peritoneal cavity. Such instability causes aberrant protein aggregation, which frequently impairs insulin delivery due to partial or complete obstruction of the pump. This problem is more severe with use of rapid-acting insulin analogs (HumalogTM (Lilly) and NovalogTM (Novo-Nordisk)), whose favorable pharmacokinetic properties otherwise offer significant advantages for use in external insulin pumps. The perfect pump insulin for an implantable system would combine rapid pharmacokinetics with long-term stability at high protein concentration as stored in a pump reservoir with gentle agitation at 37 oC. Because the propensity of insulin to misfold and undergo aberrant aggregation has seemed intrinsic to its structure, past research efforts have focused on alternative approaches: development of novel tubing materials or small-molecule surfactants to add to the insulin solution. Although some progress has been obtained, clinical success has been elusive. We propose a novel approach to the design of insulin analogs based on the topological requirements of aberrant insulin aggregation and fibrillation. The essential idea is based on our recent foundational studies of proinsulin and the mechanism of insulin fibrillation (Huang, K. et al. J. Biol. Chem. 280, 42345-55 (2005) and Huang, K. et al. Biochemistry 45, 10278-93 (2006)). These studies demonstrate a profound effect of the connecting peptide on the propensity of single-chain insulin analogs to undergo aberrant aggregation, including surface-induced fibrillation at 37 C as occurs in occluded pumps. We propose to exploit these observations to design an optimal pump insulin, designing out fibrillation while } designing in} rapid action with high insulin activity and low IGF-related mitogenicity. Design principles will be validated through a series of functional assays and molecular studies to provide a solid scientific foundation for clinical translation.